Bayesian filtering & Message Fingerprinting
Bayesian Filtering and Messaging Fingerprinting moving into Genomic Sequence Analysis. The difference between Tier 1 and the ‘affordable’ solutions.
First, we must explain Bayesian filtering and message fingerprinting:
Bayesian statistical analysis
Bayesian requires a database consisting of thousands of spam and legitimate emails, referred to as spam and ham collections. The content of each new email is analyzed and segmented into strings. These strings are compared with spam and ham collections and classified according to how often they appear in both categories. This frequency is calculated using Bayes’ theorem then a score is obtained, indicating the probability that the message is spam. Any email above a certain threshold will thus be considered spam.
Message fingerprinting
At the beginning of 2006, spammers introduced image spam, a new type of spam that effectively thwarted content analysis. Words contained in this new form of spam were incorporated into an image, making it impossible for detection systems to recognize them. This problem presented a challenge to the global anti-spam industry. Even Optical Character Recognition (OCR); a costly technique, has proved useless. So far, the only solution that can effectively prevent image spam is message fingerprinting.
Message fingerprints are profiles created from the unique form and characteristics of known spam. Message fingerprinting providers collect spam that has been reported to them from around the world by reputable volunteers. The providers use the spam to produce a message fingerprint, which is then incorporated into the filtering system in order to accurately detect messages that have a similar profile.
Why Spambrella is a Tier 1 Email Security Service Provider
Mounting an effective defense against spam requires detection techniques that evolve as quickly as the attacks themselves. Without the ability to adapt automatically to new types of threats, anti-spam defenses will always remain a step behind the spammers. Spambrella connects with Proofpoint MLX™ technology which uses advanced machine learning techniques to provide comprehensive spam detection that guards against the spam threats of today, as well as tomorrow. Proofpoint MLX continuously analyzes millions of messages and automatically adjusts its detection algorithms to identify even the newest, most cunning types of attacks. Proofpoint MLX provides accurate, adaptive, and continuous protection against spam without requiring manual tuning or administrator intervention. Spambrella is thus regarded as a tier 1 provider and is fully backed by Proofpoint Inc. in providing industry-leading (Gartner leader) technology.
Genomic Sequence Analysis
The Spambrella service employs a full range of classification methods, from legacy approaches such as heuristics and Bayesian analysis to state-of-the-art machine learning algorithms (such as those used in genomic sequence analysis) and proprietary image analysis methods. Analyzing millions of messages each day, Proofpoint MLX automatically adjusts its detection algorithms to identify even the newest spam attacks without manual tuning or administrator intervention. As a result, Proofpoint MLX is able to provide continuous spam detection and content filtering with a very high degree of accuracy–typically on the order of 99.8% or higher.
Unlike other anti-spam and malware detection solutions, the Spambrella platform delivers anti-spam defenses that do not degrade over time. By adapting continuously to the changing nature of email attacks, Spambrella ensures that Enterprises always benefit the latest anti-spam defenses–even if those defenses are only a few hours old. Spambrella threat detection technology protects corporate infrastructure against the spam and malware threats of today, as well as tomorrow.
Is this all my organization needs to be safe from Spam?
Sadly not no although this is a major step. The threat landscape changes daily which is why we highlight the importance of real-time threat detection technology and intelligence feeds. Spambrella has utilized both the Intel Security ‘Global Threat Intelligence Network (GTI) and the Proofpoint MLX which puts us in a leading position. However, there are additional layers of security we have added which you can enquire about or indeed learn from additional materials on the website.
FAQ's
Bayesian filtering is a statistical technique that analyzes the content of emails and calculates the probability that a message is spam. It compares words and patterns against large collections of known spam and legitimate emails to classify incoming messages.
Bayesian analysis breaks down email text into tokens and checks how often these appear in both spam and legitimate messages. Based on Bayes’ theorem, it assigns a score. If the score exceeds a threshold, the message is flagged as spam.
Message fingerprinting creates unique digital “signatures” of known spam campaigns. When a new email matches a known fingerprint, it is quickly identified and blocked, even if spammers change content slightly or use images.
Genomic sequence analysis applies advanced machine learning methods similar to DNA sequence comparison. It helps detect hidden patterns and evolving spam campaigns with extremely high accuracy.
Unlike basic rule-based filters, Bayesian and fingerprinting methods adapt automatically to new threats. This ensures higher accuracy, fewer false positives, and protection against advanced spam techniques such as image-based or obfuscated attacks.
Spambrella, powered by Proofpoint MLX technology, delivers adaptive spam detection with over 99% accuracy. It combines Bayesian filtering, fingerprinting, and genomic analysis, providing enterprise-level protection that cheaper, non-adaptive solutions cannot match.
Related
Related Case Studies
The service is great at filtering bad email as well as junk email out while allowing clean email though. I have used a few other options over the years and this is the best I have found. Clients sometimes have trouble configuring their settings to how they want it to be. Or tag emails as approved when they shouldn't and need IT interaction to resolve. Maybe just ease of use or having a more clear way for clients to resolve basics on their own.
Easy to onboard my customers from another spam filtering system. Very fast and haven't had any downtime in the 9 months since I have moved to Spambrella. When I have had to use support, responses where quick. I had to move all my customers from another filtering system with little notice. After I moved my customers I realised how bad the old solution I used was. Contact with Sales and Support always been professional
Used the software for: 2+ years - 5/5 Overall
With an ever overloaded department, and with cybersecurity skills shortage getting worse securing the I.T infrastructure.
Offloading the task of e-mail filtering to Spambrella has dramatically helped in the department's performance. The only drawback in our case is that the service is hosted outside of our territory and thus out of the legal jurisdiction.
I found spambrella to be easy to set up and has dramatically reduced the number of spam emails hitting our inboxes. It was easy to 'train' the software to release any genuine emails that were caught or add any spam that was not picked up.
Within a couple of weeks of use virtually no spam arrived to our mailboxes. Spam and phishing emails are a growing problem for everyone I'm sure. I now get a very low incidence of spam.
Robust, versatile, and reliable...
The reliability of the service and the level of protection that it provides. My spam levels immediately dropped to near zero.
There are almost no false positives. And I'm easily able to customize the level of protection with whitelists, blacklists, and sensitivity settings. I'm also a big fan of the antivirus and URL scanning features.
It doesn't require an arcane knowledge to set the Spam filtering up, the guides are straight to the point and support staff are very helpful. Functionality wise, in short: we do not get spammed. Thanks to Spambrella.
Archiving wise, the new solution is easy to use, searches well and fast and is by far the cheapest we could find at the time. Ten year retention rocks!
